In this revised application, we propose to measure the effects of amphetamine challenge vs. placebo on mood and regional cerebral blood flow (rCBF) in patients with RDC endogenous Major Depressive Disorder and in age-and gender-matched normal controls. The fundamental purpose of this investigation is to identify CNS regions that subserve or participate in mood alterations induced by amphetamine in normal controls and depressed patients. Our primary aims are: 1) to identify and characterize changes in three-dimensional distribution of rCBF induced by administration of amphetamine compared to placebo; 2) to determine relationships between amphetamine-induced changes in rCBF and changes in mood; 30 to determine if amphetamine-induced rCBF changes distinguish between normals and depressed patients; and 4) to advance the development of statistical analysis methods for three-dimensional functional brain imaging data, with specific applications for analyzing challenge effects and between-group differences in rCBF. Secondary aims are; 50 to determine whether there are relationships between the rCBF response to amphetamine and clinical features of depression; 6) to determine whether there are relationships between resting rCGF abnormalities and clinical features of depression; and 7) to determine if high-resolution three-dimensional images of resting rCBF replicates reported differences between normal controls and depressed patients. We originally proposed to study both the endogenous and nonendogenous subtypes of Major Depression because they may differ in symptom pattern, response to treatment, age distribution, biological abnormalities (e.g. DST, sleep polysomnography), and in patterns of CNS functional abnormalities. However, our study section recommended that we omit nonendogenous subjects to minimize sources of variability in our data. We concur with this advice, and the proposal has been so modified. Two groups of 80 subjects each will be studied: normal controls and patients with endogenous depression. The experiment has a fixed-order design in which subjects receive placebo then amphetamine (0.4 mg/kg p.o). The experimental plan provides three major opportunities; 1)to test hypotheses based on existing animal data and our pilot data regarding the location and nature of amphetamine-induced rCBF changes (such hypotheses address a delineation of the functional neuroanatomy of amphetamine-induced mood changes); 2) to provide original data for hypothesis generation as to whether mood-associated systems differ functionally and/or regionally among the groups under study; and 3) to develop and test innovative sophisticated statistical approaches to complex but potentially powerful imaging data in a carefully evaluated population.